Revisiting fiscal federalism: Implications of Illinois’s and Chicago’s fiscal problems
Broadly speaking, fiscal federalism is a theory of public finance concerned with how to most appropriately and efficiently provide government services (or public goods) through different levels of government. This theory also involves how to set up the fiscal relationships (e.g., conditional versus unconditional transfers) between the different levels of governments to best provide these services. In the United States (and many other countries), there have traditionally been three levels of government—federal, state, and local—among which to divide up key government functions. According to the purist form of the theory, decentralization away from the federal government promotes welfare gains, as the scale of provision of particular services is scaled to the size of the population being served. This is partly based on the idea that public goods should be defined by geography, such that there are national public goods (such as defense) and local public goods (such as public school systems). Moreover, cost efficiencies—and benefit spillovers—may occur when the level of public service is calibrated to the particular preferences of the electorate in that geography rather than the central government providing the service in a uniform way.1 Generally, the theory of fiscal federalism has been a guiding principle for the design and delivery of government services in the United States.
In 2008, public finance economist Wallace Oates suggested a revised theory for fiscal federalism.2 Oates observed that while lower levels of government (states or municipalities) may have explicit rules against running budget deficits or amassing unsupportable levels of debt, they often ignore these restrictions under the hope and belief that a higher level of government (the federal government or states) will bail them out. For Oates, the lower level of government’s political incentives for not having to make difficult fiscal adjustments might outweigh those for exercising fiscal prudence. Therefore, lower levels of government can engage in fiscal behavior that can place burdens on higher levels of government. This reallocation of fiscal costs can undermine the efficiencies gained under the traditional notion of fiscal federalism. In this blog entry, I will consider fiscal federalism, including Oates’s revised version, in light of Illinois’s and Chicago’s recent fiscal challenges. Is the state or the city skirting fiscal rules in the hopes that a higher level of government might bail it out? Are the political incentives not to solve the problems now large enough to forestall any further action to fix them?
New perspectives on fiscal federalism
Oates suggests an important revision to thinking about the actual operation of fiscal federalism. Oates argues that rather than promoting the efficient provision of government services, fiscal federalism can incentivize certain levels of government to try to extract or indeed extract (albeit inefficiently) resources from other levels of government. An example of this is a “soft-budget constraint.” If there’s an implicit understanding that a lower level of government can count on help from a higher level of government when the former gets into fiscal difficulties, the lower level may be encouraged to unduly run deficits and expand debt. Essentially, for the lower level of government, there’s no credibility to any pledge by the higher level of government to not intercede during a time of fiscal stress. The lower level of government operates under the assumption that the higher level of government is interested in the welfare of all of its citizens, so it cannot allow a lower level of government to fail. The key question is why there is a soft-budget constraint? One theory is that the fiscal responsibilities between governments are poorly defined, which can lead to fiscal misbehavior. For instance, local governments may view themselves as providing essential public services that higher levels of government could justify supporting during an economic downturn. So, local governments may not be maintaining a rainy day fund, instead allocating dollars that should have reserved for such a fund to other parts of the budget that are more politically expedient. A second theory is that in many cases, the bond market presently does not—or perhaps cannot—accurately account for the political or financial risk of some government bonds. This may be reflected in fairly positive bond ratings for poorly fiscally performing units of government. Of course, bond ratings are designed to reflect the default risk attached to a specific bond issuance and not necessarily the underlying strength of the issuing government. If the bond covenant provides protections and preferences for the bond to be repaid even in the face of poor fiscal performance, the rating will reflect this. Still, such “mispricing” in the bond market can allow for mischief on the part of lower levels of government. So, for example, if borrowing (to cover a deficit) can continue even in the face of fiscal instability, the only penalty a profligate local government faces is having to pay a higher borrowing rate. Similarly, if the capitalization of poor fiscal performance into land values is not readily recognized (or in the case of strong fiscal performance, not rewarded), the political penalties of poor management may be small, at least to the current stewards. Additionally, if there is a history of bailouts by the federal government during recessions, it can be rational to assume that lower levels of government can be bailed out by a higher level of government that does not face a budget constraint. This belief can breed fiscal misbehavior among lower levels of government.
Considering the fiscal woes of the State of Illinois and the City Chicago, there’s a strong case to be made that all of the aforementioned may have occurred. First, despite frequent and substantial credit rating downgrades, both Illinois and Chicago have repeatedly issued bonds despite their deteriorating fiscal conditions. While these bonds have carried higher interest rates, the ability of the state and local government to take on more debt despite their fiscal problems suggests that the bond market provides little discipline for poor fiscal behavior. Second, Chicago has sought fiscal relief for times when the state has tried to impose a fiscal limit. The state passed new pension contribution requirements for Chicago; and when the costs of funding these requirements became apparent, the city government requested a new schedule for making these payments. Even at the state level, the funds that Illinois received from the federal government in the wake of the Great Recession (as part of the countercyclical aid program instituted under the American Recovery and Reinvestment Act of 2009) arguably helped the state not make fiscal adjustments in light of deteriorating circumstances. In practice, this means that residents potentially overconsume public services because they do not pay the full tax associated with providing the service.
How to make fiscal decentralization more effective
With all that said, this discussion of the theory of fiscal federalism can still extend to whether the “right” level of government is delivering the “right” service and which level of government should be paying for that service. Early childhood education is a good example illustrating the challenges associated with figuring these things out. Often early childhood education is provided at the local level (sometimes, it’s provided at the state level); yet given the national returns to improving educational outcomes, some would argue that this should be a federally funded program.
To determine the size of welfare gains from fiscal decentralization, it is important to calculate the variation in demand for the government services across jurisdictions, as well as the variation in costs for providing these services. If no variation existed, central provision of a uniform public service would probably make the most sense. However, since variation does exist, welfare gains can be realized, assuming the state or local government can estimate the level of service that residents demand. Under fiscal federalism, a multilevel system of government efficiently provides different types of public services to its specific constituencies. A complicating factor is when a government service provides a spillover benefit to another constituency. In a case of a locally provided government service that has a spillover benefit, an efficient form of fiscal federalism would suggest that the higher level of government provide a grant in order to encourage the local government to provide the service in a socially efficient way. An example of this might be a grant from the federal government or state government to a local police force to share its law-enforcement database with others. Oates also argues that in some cases fiscal equalization grants might be called for when a locality lacks the tax base and/or resources or faces high costs in providing particular services.
A new model for fiscal federalism?
Coupled with the original notion of a fiscal federalist system, Oates’s critique suggests a new model may be needed. This model would have two components. First, there would be a binding budget constraint on both the state and local governments. If policymakers in both subnational governments knew that they had to provide truly balanced budgets on an annual basis, they could not create budget deficits under the assumption that their governments could be bailed out by a higher level of government. In practice this would mean that each subnational government would have to have a structurally balanced budget,3 based on normal trends in the state or locality. Work by Richard Dye and David Merriman4 has created a structural model of total state expenditures and revenues for the State of Illinois, which allows the future budget performance to be projected under the assumption that the underlying trends in expenditures and revenues are maintained. This type of model allows an estimation of the structural budget gap between expenditures and revenues to be identified.
To be clear, a binding budget constraint would need to exist to determine the structural trend in the state or localities budget. The exception where aid from the federal government could be warranted is in the case of a national recession. During a downturn in the business cycle, maintaining social service and other countercyclical state and local spending can have a positive spillover effect to the national economy and, therefore, reflect good fiscal policy. However, even in this case, the federal assistance should be linked to an objective, rules-based method for determining the timing and appropriate level of federal support.5 The alternative to permitting federal intervention during a national recession would be to require states to carry larger budget reserves to smooth out what would otherwise be volatile fiscal behavior during a downturn. The size of the reserves could be determined from a stress-test type model, where the sensitivity of the state budget to differing economic scenarios could be estimated. States with more stable revenues and expenditures in fiscal downturn scenarios would carry smaller reserves than those with more volatile taxes and spending.6 Again, the approach would be to base this on a rules-based system to ensure all states and localities comply. Part of the goal of such a binding system is to improve fiscal transparency: This approach would make the true tax price of providing government services readily apparent.
The second element of this revised federalist model is to do a better job at estimating where spillovers occur in government provision to ensure that the right level of government is providing resources for the service. Robert Inman7 has consistently argued that it is inefficient for cities—which often have larger shares of distressed population than suburbs and rural areas—to be responsible for funding government programs targeted to this population. His preferred strategy would be for the city government to develop and administer the government programs, but with funding provided at the regional level. Under the usual fiscal federalist structure, the city absorbs the cost of providing a service (to the distressed population) that yields regional spillovers. Inman has further argued that program provision at the local level can be more efficient because those in the local government will likely know the needs of the population better than those in a higher level of government. He suggests that this is what occurs in Pittsburgh: The county funds Pittsburgh’s social welfare programs, and the city is responsible for administering them. When cities are forced to bear these costs alone, this can limit their ability to fund other government services that might encourage private sector investment. An example of a more controversial extension of this idea would be to examine how investments in human capital and education are made. If it can be assumed that there is a national economic return to human capital investment, an education model that is based on local preferences may not be the most efficient. A purely local model would make sense if it can be assumed that the level of education provided meets local taxpayers’ expectations and that recipients of the education stay in the locality. However, under the assumption that promoting labor mobility is desirable, changes in certain expectations for education might be appropriate. For example, state reductions in funding for higher education has been frequently lamented. A clear case for state support could in theory be based on whether the graduates stay in the state after graduation so that the state and taxpayers receive a return on their education investment. Assuming that students at community or regional institutions within the state may be less mobile, state support may be more justified for these types of schools than for a flagship university drawing an increasingly national or international student body that is more mobile after graduation. In fact, in most cases, recent reductions in support for public university systems have been most pronounced for the flagship public institutions. That said, if the spillovers from the national public universities are to the nation as a whole, this suggests that increased federal support might be appropriate. Again, the goal is to identify where the spillover occurs and to better match where the funding comes from with the level of government benefiting from the service provision.
Conclusion
Fiscal federalism’s guiding principles have generally served the United States well throughout its history. However, Oates and others are correct in arguing that these principles, along with the underlying theory, need to be revised—especially given how federal, state, and local governments have behaved over the past few years.
Footnotes
1 Another gain from fiscal decentralization may be innovation. Allowing service delivery experimentation tailored to the constituency the level of government is serving can lead to breakthroughs in service delivery, which may applied to other jurisdictions.
2 Wallace E. Oates, 2008, “On the evolution of fiscal federalism: Theory and institutions,” National Tax Journal, Vol. 61, No. 2, June, pp. 313–334, available online.
3 To see what I mean by “structurally balanced budget,” see this link.
4 See, for instance, this link.
5 For a discussion on how to optimally design federal support, see Richard H. Mattoon, Vanessa Haleco-Meyer, and Taft Foster, 2010, “Improving the impact of federal aid to the states,” Economic Perspectives, Federal Reserve Bank of Chicago, Vol. 34, Third Quarter, pp. 66–82, available online.
6 Richard Mattoon, 2003, “Creating a national state rainy day fund: A modest proposal to improve future state fiscal performance,” Proceedings: Annual Conference on Taxation and Minutes of the Annual Meeting of the National Tax Association, Vol. 96, pp. 118–124, available online.
7 Robert P. Inman, 1995, “How to have a fiscal crisis: Lessons from Philadelphia,” American Economic Review, Vol. 85, No. 2, May, pp. 378–383, available online.
